Montreal Music and Machine Learning Workshop 2008

Overview

This will be a one-day workshop in Montreal on the topic of machine learning and music. It will be held on November 14th at the University of Montreal, with both morning and afternoon sessions. The target audience of this workshop includes researchers from fields such as machine learning, music information retrieval, music technology, cognitive psychology, neuroscience, and motion capture. You need not be a machine learning expert to attend!

Note: The workshop is now full. However there may be cancellations. If you would like to attend please contact Douglas Eck (douglas [dot] eck [at] umontreal [point] ca)

Thanks from the organizers,
Douglas Eck (Université de Montréal Computer Science)
Michael Mandel (Columbia University Electrical Engineering)
Pascal Vincent (Université de Montréal Computer Science)

Schedule

Morning Session (9:00 to 12:00)
9:00 Machine Learning and the Curse of Highly-Variable Functions (Yoshua Bengio, Univ. de Montréal Comp. Sci.)
9:30 Deep Componential Models for Human Motion (Graham Taylor, U. of Toronto Comp. Sci.)
10:30 Coffee Break + Discussion
11:00 Stairway to Muskrat Love - If you like Led Zeppelin, you might like Captain & Tennille (Paul Lamere, Sun Labs)
11:30 Mapping Music onto Words (Douglas Eck, Univ. de Montréal Comp. Sci./BRAMS/CIRMMT)
 
Lunch / After-Lunch Demonstration (12:00 to 1:00)
12:00 Lunch (provided)
12:30 Demo: Automatic expressive performanceon Boesendorfer piano (Stanislas Lauly, Univ. de Montréal Comp. Sci./BRAMS/CIRMMT)
 
Afternoon Session (1:00 to 4:00)
1:00 Shift-Invariant Sparse Coding for Audio Classification (Roger Grosse, MIT CSAIL)
2:00 A Summary of Recent Advances in Optical Recognition for Early Music Documents (John Ashley Burgoyne, McGill Music Tech./CIRMMT)
2:30 Coffee Break + Discussion
3:00 Proximity Graphs for Instance-Based Learning (Godfried Toussaint, McGill Computer Science/CIRMMT)
3:30 A Probabilistic Model for Melodies (Jean-François Paiement, Yahoo! Labs)
 
Poster and Demo Sesson (4:00 to 6:00) / Wine and Cheese(5:00 to 6:00)
4:00James Bergstra (Université de Montréal Computer Science/BRAMS/CIRMMT)
 Thierry Bertin-Mahieux (Columbia University Engineering/BRAMS/CIRMMT)
 Johanna Devaney (McGill Music Tech/CIRMMT)
 Andrew Hankinson (McGill Music Tech/CIRMMT)
 Jason Hockman (McGill Music Tech/CIRMMT)
 Steven Livingstone (McGill Psychology)
 François Maillet (Université de Montréal Computer Science/BRAMS/CIRMMT)
 Michael Mandel (Columbia University Engineering), "MajorMiner: automatically describing music"
 Pierre-Antoine Manzagol (Université de Montréal Computer Science), "On the Use of Sparse Time-Relative Auditory Codes for Music"
 Cory McKay (McGill Music Tech/CIRMMT)
5:00Wine and Cheese; Poster And Demo Session continues

Location

The workshop will be held at the International Laboratory for Brain Music and Sound (BRAMS located at 1430 Boulevard Mont-Royal 2nd Floor.) When you arrive at 1430 BOulevard Mont-Royal take the door marked "1430", go up one flight of steps. The door will be on your right. Nearest metro is Édouard Montpetit (follow exit for Faculté de Musique). See www.brams.org/contact.html for more specific instructions.


View Larger Map

Organization and Sponsors

This workshop is organized by the CIRMMT Axis on Music Information Retrieval and and by the LISA Machine Learning Lab. Space and technical support are provided by BRAMS. Additional funding for this workshop is provided by MITACS.

Centre for Interdisciplinary Research in Music Media and Technology (CIRMMT) Mathematics of Information Technology and Complex Systems (MITACS) International Laboratory for Brain Music and Sound (BRAMS)